Re: Conlang Unicode Font (was Re: Kamakawi Unicode Font Question)

On 08/03/08 16:45:50, Eric Christopherson wrote:
>
> On Mar 7, 2008, at 5:36 AM, Tristan McLeay wrote:
> >
> > The insuperable problem is that unfortunately OpenType, the main
> > modern
> > font format, doesn't support all its features everywhere. Automatic
> > ligatures apparently don't happen if the unligated characters are
> in
> > the PUA. Another problem is that Latin handwriting fonts can't
> > use context-sensitive glyphs the way Arabic fonts do (for instance,
> my
> > cursive p and b are open when a letter follows, but closed when
> they
> > don't --- you can't tell an OpenType font to automatically do
> that).
>
> How is it that such a font can produce ligatures when e.g. <f> is
> followed by <i> or <fl>, but not vary the look of <p> and <b>
> depending on what follows?
The difference is that fi and ffl form ligatures, so we combine both
the first and second characters into a new one. With open-versus-closed
p, we're replacing only the p based on the context; we want the open
variant if any Latin alphabetic character follows, and closed p
otherwise.

However, as to the details I provided in the last post, I might have
been slightly wrong in that discussion. From this page:
  <http://fontforge.sourceforge.net/editexample6-5.html#Conditional>
it would seem that some OpenType implementations support conditional
variants in all scripts, and others don't. If I understand it, the
standard doesn't require it, but there's nothing to stop a complying
implementation from supporting it, other than lack of imagination on
the implementer's part.

--
Tristan.